Prognostic and predictive value of tumor infiltration proportion within lymph nodes in N1 colorectal cancer

TL;DR

This study shows that the amount of tumor spread into lymph nodes can predict survival in colorectal cancer patients and improve treatment decisions.

Contribution

The study introduces a new TIPLN-based nomogram that improves survival prediction accuracy for N1 colorectal cancer patients.

Findings

High TIPLN levels are significantly linked to poorer survival in N1 colorectal cancer patients.

The TIPLN-based nomogram outperformed traditional clinicopathological models in predicting survival outcomes.

The nomogram demonstrated strong precision and clinical utility in both training and validation cohorts.

Abstract

Lymph node metastasis is a crucial determinant of prognosis in colorectal cancer (CRC), significantly impacting survival outcomes and treatment decision-making. This study aims to evaluate the prognostic value of tumor infiltration proportion within lymph nodes (TIPLN) in N1 CRC patients and to develop a TIPLN-based nomogram to predict prognosis.

A total of 416 N1 CRC patients who underwent radical resection were enrolled and divided into training and validation cohorts. Whole-slide images of lymph nodes were annotated to assess the TIPLN. Univariable and multivariable Cox regression analyses were conducted to identify independent prognostic factors and to develop a nomogram for predicting patient outcomes. The precision and discrimination of the nomogram were evaluated using the area under the receiver operating characteristic curve (AUC), concordance index (C-index), and calibration curve. Decision curve analysis (DCA) was performed to compare the net benefit of the nomogram at different threshold probabilities. Additionally, net reclassification index (NRI) and integrated discrimination improvement (IDI) were used to evaluate the nomogram’s clinical utility.

High TIPLN levels were significantly associated with poorer overall survival (OS). Five variables, including TIPLN, were selected to construct the nomogram. The C-index in OS prediction was 0.739 and 0.753 for the training and validation cohorts, respectively. Additionally, strong precision and discrimination were demonstrated through AUC and calibration curves. The NRI (training cohort: 0.191 for 3-year and 0.436 for 5-year OS prediction; validation cohort: 0.180 for 3-year and 0.439 for 5-year OS prediction) and IDI (training cohort: 0.079 for 3-year and 0.094 for 5-year OS prediction; validation cohort: 0.078 for 3-year and 0.098 for 5-year OS prediction) suggest that the TIPLN-based nomogram significantly outperformed the clinicopathological nomogram. Furthermore, DCA demonstrated the high clinical applicability of the TIPLN-based nomogram for predicting OS.

TIPLN could serve as a prognostic predictor for N1 CRC patients. The TIPLN-based nomogram enhances survival prediction accuracy and facilitates more informed, individualized clinical decision-making.

Figures